View Single Post
Old 01-11-2014, 09:30 AM   #22 (permalink)
Registered Users
Posts: 1,667
Thread Starter Thread Starter

Join Date: Mar 2013


All servo outputs from the 130x 3-in-1 use a Pulse Width Modulation (PWM) signal to indicate the desired position. Servo center position is commanded through a 1520 microsecond (uSec) wide pulse. In addition to the variable wiper position, the servo microcontroller can read the voltage for a fixed tap on the servo resistive strip. The servo appears to locate the actuator above this fixed tap when it is commanded to the center position.
NOTE: A method for visually verifying the center position of a 130x linear servo has been added to Part 3, Inspections and Repairs.
With a DX6i set to D/R of 100%, the pulse width from the 3-in-1 was found to vary from around 1275 uSec to 1840 uSec with maximum TX stick movements. A shorter pulse width drives the actuator towards the gear end of the servo. A wider pulse width drives the actuator towards the brass locknuts on the threaded rod of the servo mechanics.
The signal from the 3-in-1 has a pulse repetition rate of 450 Hz.
Pulses applied to the servo motor are not synchronized to the timing of the 3-in-1 signals. The servo microcontroller appears to have the capability of generating a motor pulse about every 1.4 millisecond (mSec). In each 1.4 mSec window, no pulse will be generated if no movement of the servo position is necessary. If movement is desired, the microcontroller will provide a motor drive pulse with a width from between about 40 uSec to about 1 mSec. Most motor pulses observed in at-rest jitter were between 50 uSec and 100 uSec wide. The microcontroller will control the direction of the motor rotation achieved by applying pulses as required to two inputs of the motor drive circuit.
No quantifiable differences were observed between a linear tail servo and a linear cyclic servo.
Information presented in this post is based on signals observed on a 130x at rest using an analog oscilloscope without trace storage or transition measurement capability. The varying nature of the motor drive pulse train is in particular hard to characterize properly without at least single-sweep trace storage. Data provided here should be considered notional, and likely does NOT represent worst case values for real-world pulse width minimums and maximums.
Kevin B.
"Fun, Learning, Friendship and Mutual Respect" - good words to live by
RC: 300x, 130x, DX6i, iCharger 106b+, Phoenix Sim
Other: Atmel AVR microcontrollers, electronics, 3D printing

Last edited by helibus; 01-13-2014 at 04:58 PM.. Reason: missing paragraph breaks... Grrr.
helibus is offline        Reply With Quote